Open Software Development Research Articles

Assessing the Accuracy of Artificial Intelligence Models in Scoliosis Classification and Suggested Therapeutic Approaches.

Background: Open-source artificial intelligence models (OSAIMs) are increasingly being applied in various fields, including IT and medicine, offering promising solutions for diagnostic and therapeutic interventions. In response to the growing interest in AI for clinical diagnostics, we evaluated several OSAIMs-such as ChatGPT 4, Microsoft Copilot, Gemini, PopAi, You Chat, Claude, and the specialized PMC-LLaMA 13B-assessing their abilities to classify scoliosis severity and recommend treatments based on radiological descriptions from AP radiographs. Methods: Our study employed a two-stage methodology, where descriptions of single-curve scoliosis were analyzed by AI models following their evaluation by two independent neurosurgeons. Statistical analysis involved the Shapiro-Wilk test for normality, with non-normal distributions described using medians and interquartile ranges. Inter-rater reliability was assessed using Fleiss' kappa, and performance metrics, like accuracy, sensitivity, specificity, and F1 scores, were used to evaluate the AI systems' classification accuracy. Results: The analysis indicated that although some AI systems, like ChatGPT 4, Copilot, and PopAi, accurately reflected the recommended Cobb angle ranges for disease severity and treatment, others, such as Gemini and Claude, required further calibration. Particularly, PMC-LLaMA 13B expanded the classification range for moderate scoliosis, potentially influencing clinical decisions and delaying interventions. Conclusions: These findings highlight the need for the continuous refinement of AI models to enhance their clinical applicability.

WebGIS Open-source Platform for Localizations of New P2G Plants in Sicily

The climatic emergency that involves the globe has led targets of greenhouse gas reduction in the EU and all over the world. In this scenario, recent advances in renewable renewable energy sources (RESs) have focused interest on the diffusion of power supplies that are produced by photovoltaic and wind plants. The non-programmable nature of these energy sources has led recent studies to consider the power-to-gas (P2G) solution as an opportunity for employing the curtailed electric energy by converting it into hydrogen. The localizations of P2G plants depend on several factors regarding power production, distances, and population distributions. The necessity of integrating these factors led this work to study the development of a cost function that is hosted by a webbased GIS (geographic information system) platform, thus allowing for the storage, elaboration, and web fruition of an entire data set that is related to the possible new localizations of P2G plants. The structure is based on open-source technology and creates a solution that is easily employable by specialists. The developed platform is composed of different remotely connected blocks that are solely based on open-source technology and is focused the interest on the territory of Sicily (Italy). GIS software, a RDBMS database, a geospatial server (Geoserver ), a Python optimization module, and a WebGIS visualizer are integrated. This work represents a scientific contribution to the management of energy sources, with a particular focus on policies that are based on hydrogen technology. In fact, different data sets that contain several levels of information that are related to the management and the localization of P2G plants will be even further employed in the future.

Operational Forest-Fire Spread Forecasting Using the WRF-SFIRE Model

In the present research, the open-source WRF-SFIRE model has been used to carry out surface forest fire spread forecasting in the North Sikkim region of the Indian Himalayas. Global forecast system (GFS)-based hourly forecasted weather model data obtained through the National Centers for Environmental Prediction (NCEP) at 0.25 degree resolution were used to provide the initial conditions for running WRF-SFIRE. A landuse–landcover map at 1:10,000 scale was used to define fuel parameters for different vegetation types. The fuel parameters, i.e., fuel depth and fuel load, were collected from 23 sample plots (0.1 ha each) laid down in the study area. Samples of different categories of forest fuels were measured for their wet and dry weights to obtain the fuel load. The vegetation specific surface area-to-volume ratio was referenced from the literature. The atmospheric data were downscaled using nested domains in the WRF model to capture fire–atmosphere interactions at a finer resolution (40 m). VIIRS satellite sensor-based fire alert (375 m spatial resolution) was used as ignition initiation point for the fire spread forecasting, whereas the forecasted hourly weather data (time synchronized with the fire alert) were used for dynamic forest-fire spread forecasting. The forecasted burnt area (1.72 km2) was validated against the satellite-based burnt area (1.07 km2) obtained through Sentinel 2 satellite data. The shapes of the original and forecasted burnt areas matched well. Based on the various simulation studies conducted, an operational fire spread forecasting system, i.e., Sikkim Wildfire Forecasting and Monitoring System (SWFMS), has been developed to facilitate firefighting agencies to issue early warnings and carry out strategic firefighting.

HPVsim: An agent-based model of HPV transmission and cervical disease.

In 2020, the WHO launched its first global strategy to accelerate the elimination of cervical cancer, outlining an ambitious set of targets for countries to achieve over the next decade. At the same time, new tools, technologies, and strategies are in the pipeline that may improve screening performance, expand the reach of prophylactic vaccines, and prevent the acquisition, persistence and progression of oncogenic HPV. Detailed mechanistic modelling can help identify the combinations of current and future strategies to combat cervical cancer. Open-source modelling tools are needed to shift the capacity for such evaluations in-country. Here, we introduce the Human papillomavirus simulator (HPVsim), a new open-source software package for creating flexible agent-based models parameterised with country-specific vital dynamics, structured sexual networks, and co-transmitting HPV genotypes. HPVsim includes a novel methodology for modelling cervical disease progression, designed to be readily adaptable to new forms of screening. The software itself is implemented in Python, has built-in tools for simulating commonly-used interventions, includes a comprehensive set of tests and documentation, and runs quickly (seconds to minutes) on a laptop. Performance is greatly enhanced by HPVsim's multiscale modelling functionality. HPVsim is open source under the MIT License and available via both the Python Package Index (via pip install) and GitHub (hpvsim.org).

Advancements in hand-drawn chemical structure recognition through an enhanced DECIMER architecture

Accurate recognition of hand-drawn chemical structures is crucial for digitising hand-written chemical information in traditional laboratory notebooks or facilitating stylus-based structure entry on tablets or smartphones. However, the inherent variability in hand-drawn structures poses challenges for existing Optical Chemical Structure Recognition (OCSR) software. To address this, we present an enhanced Deep lEarning for Chemical ImagE Recognition (DECIMER) architecture that leverages a combination of Convolutional Neural Networks (CNNs) and Transformers to improve the recognition of hand-drawn chemical structures. The model incorporates an EfficientNetV2 CNN encoder that extracts features from hand-drawn images, followed by a Transformer decoder that converts the extracted features into Simplified Molecular Input Line Entry System (SMILES) strings. Our models were trained using synthetic hand-drawn images generated by RanDepict, a tool for depicting chemical structures with different style elements. A benchmark was performed using a real-world dataset of hand-drawn chemical structures to evaluate the model's performance. The results indicate that our improved DECIMER architecture exhibits a significantly enhanced recognition accuracy compared to other approaches.Scientific contributionThe new DECIMER model presented here refines our previous research efforts and is currently the only open-source model tailored specifically for the recognition of hand-drawn chemical structures. The enhanced model performs better in handling variations in handwriting styles, line thicknesses, and background noise, making it suitable for real-world applications. The DECIMER hand-drawn structure recognition model and its source code have been made available as an open-source package under a permissive license.Graphical

Position paper: Open-source technology in the treatment of people living with diabetes mellitus-an Austrian perspective : Technology Committee of the Austrian Diabetes Association

People living with diabetes mellitus can be supported in the daily management by diabetes technology with automated insulin delivery (AID) systems to reduce the risk of hypoglycemia and improve glycemic control as well as the quality of life. Due to barriers in the availability of AID-systems, the use and development of open-source AID-systems have internationally increased. This technology provides a necessary alternative to commercially available products, especially when approved systems are inaccessible or insufficiently adapted to the specific needs of the users. Open-source technology is characterized by worldwide free availability of codes on the internet, is not officially approved and therefore the use is on the individual's own responsibility. In the clinical practice a lack of expertise with open-source AID technology and concerns about legal consequences, lead to conflict situations for health-care professionals (HCP), sometimes resulting in the refusal of care of people living with diabetes mellitus. This position paper provides an overview of the available evidence and practical guidance for HCP to minimize uncertainties and barriers. People living with diabetes mellitus must continue to be supported in education and diabetes management, independent of the chosen diabetes technology including open-source technology. Check-ups of the metabolic control, acute and chronic complications and screening for diabetes-related diseases are necessary and should be regularly carried out, regardless of the chosen AID-system and by a multidisciplinary team with appropriate expertise.

New IPECAD open-source model framework for the health technology assessment of early Alzheimer’s disease treatment: development and use cases

Reimbursement decisions for new Alzheimer's disease (AD) treatments are informed by economic evaluations. An open-source model with intuitive structure for model cross-validation can support the transparency and credibility of such evaluations. We describe the new International Pharmaco-Economic Collaboration on Alzheimer's Disease (IPECAD) open-source model framework (version 2) for the health-economic evaluation of early AD treatment and use it for cross-validation and addressing uncertainty. A cohort state-transition model using a categorized composite domain (cognition and function) was developed by replicating an existing reference model and testing it for internal validity. Then, features of existing Institute for Clinical and Economic Review (ICER) and Alzheimer's Disease Archimedes Condition-Event Simulator (AD-ACE) models assessing lecanemab treatment were implemented for model cross-validation. Additional uncertainty scenarios were performed on choice of efficacy outcome from trial, natural disease progression, treatment effect waning and stopping rules, and other methodological choices. The model is available open-source as R code, spreadsheet, and web-based version via https://github.com/ronhandels/IPECAD. In the IPECAD model incremental life-years, quality-adjusted life-years (QALY) gains and cost savings were 21% to 31% smaller compared with the ICER model and 36% to 56% smaller compared with the AD-ACE model. IPECAD model results were particularly sensitive to assumptions on treatment effect waning and stopping rules and choice of efficacy outcome from trial. We demonstrated the ability of a new IPECAD open-source model framework for researchers and decision makers to cross-validate other (Health Technology Assessment submission) models and perform additional uncertainty analyses, setting an example for open science in AD decision modeling and supporting important reimbursement decisions.

Generalizable clinical note section identification with large language models.

Clinical note section identification helps locate relevant information and could be beneficial for downstream tasks such as named entity recognition. However, the traditional supervised methods suffer from transferability issues. This study proposes a new framework for using large language models (LLMs) for section identification to overcome the limitations. We framed section identification as question-answering and provided the section definitions in free-text. We evaluated multiple LLMs off-the-shelf without any training. We also fine-tune our LLMs to investigate how the size and the specificity of the fine-tuning dataset impacts model performance. GPT4 achieved the highest F1 score of 0.77. The best open-source model (Tulu2-70b) achieved 0.64 and is on par with GPT3.5 (ChatGPT). GPT4 is also found to obtain F1 scores greater than 0.9 for 9 out of the 27 (33%) section types and greater than 0.8 for 15 out of 27 (56%) section types. For our fine-tuned models, we found they plateaued with an increasing size of the general domain dataset. We also found that adding a reasonable amount of section identification examples is beneficial. These results indicate that GPT4 is nearly production-ready for section identification, and seemingly contains both knowledge of note structure and the ability to follow complex instructions, and the best current open-source LLM is catching up. Our study shows that LLMs are promising for generalizable clinical note section identification. They have the potential to be further improved by adding section identification examples to the fine-tuning dataset.

A Causal Model for Primary Prevention of Cardiovascular Disease: The Health Economic Model for the Primary Prevention of Cardiovascular Disease

ObjectivesOur objective was to design and develop an open-source model capable of simulating interventions for primary prevention of cardiovascular disease (CVD) that incorporated the cumulative effects of risk factors (eg, cholesterol years or blood-pressure years) to enhance health economic modeling in settings which clinical trials are not possible. MethodsWe reviewed the literature to design the model structure by selecting the most important causal risk factors for CVD—low-density lipoprotein-cholesterol (LDL-C), systolic blood pressure (SBP), smoking, diabetes, and lipoprotein (a) (Lp[a])—and most common CVDs—myocardial infarction and stroke. The epidemiological basis of the model involves the simulation of risk factor trajectories, which are used to modify CVD risk via causal effect estimates derived from Mendelian randomization. LDL-C, SBP, Lp(a), and smoking all have cumulative impacts on CVD risk, which were incorporated into the health economic model. The data for the model were primarily sourced from the UK Biobank study. We calibrated the model using clinical trial data and validated the model against the observed UK Biobank data. Finally, we performed an example health economic analysis to demonstrate the utility of the model. The model is open source. ResultsThe model performed well in all validation tests. It was able to produce interpretable and plausible (consistent with expectations of the existing literature) results from an example health economic analysis. ConclusionsWe have constructed an open-source health economic model capable of incorporating the cumulative effect of LDL-C (ie, cholesterol years), SBP (SBP-years), Lp(a), and smoking on lifetime CVD risk.

Self-Supervised Learning Improves Accuracy and Data Efficiency for IMU-Based Ground Reaction Force Estimation.

Recent deep learning techniques hold promise to enable IMU-driven kinetic assessment; however, they require large extents of ground reaction force (GRF) data to serve as labels for supervised model training. We thus propose using existing self-supervised learning (SSL) techniques to leverage large IMU datasets to pre-train deep learning models, which can improve the accuracy and data efficiency of IMU-based GRF estimation. We performed SSL by masking a random portion of the input IMU data and training a transformer model to reconstruct the masked portion. We systematically compared a series of masking ratios across three pre-training datasets that included real IMU data, synthetic IMU data, or a combination of the two. Finally, we built models that used pre-training and labeled data to estimate GRF during three prediction tasks: overground walking, treadmill walking, and drop landing. When using the same amount of labeled data, SSL pre-training significantly improved the accuracy of 3-axis GRF estimation during walking compared to baseline models trained by conventional supervised learning. Fine-tuning SSL model with 1-10% of walking data yielded comparable accuracy to training baseline model with 100% of walking data. The optimal masking ratio for SSL is 6.25-12.5%. SSL leveraged large real and synthetic IMU datasets to increase the accuracy and data efficiency of deep-learning-based GRF estimation, reducing the need for labeled data. This work, with its open-source code and models, may unlock broader use cases of IMU-driven kinetic assessment by mitigating the scarcity of GRF measurements in practical applications.

Modeling Electrode Shape Changes in Electrodeposition and Electrochemical Dissolution

In the fields of metal production, surface finishing, and electronic device manufacturing, achieving precise control over electrochemical processes is crucial for product quality, efficiency, and cost-effectiveness. This paper introduces an open-source modeling tool based on a dynamic mesh technique built on the finite volume method using the C ++ toolbox OpenFOAM. This tool has been validated against both newly proposed analytical solutions and existing numerical and experimental results in various conditions and kinetic controls. The analytical solution predicts the electrode surface position, current, or cell voltage difference for confined electrodes. It considers primary and secondary current distributions with linear kinetics under different electric control modes. Notably, the validation highlights the congruence of the new method with prior studies and underscores its potential to offer enhanced predictive capabilities. Furthermore, this work extends beyond traditional modeling approaches by incorporating pulse reverse plating, which has been successfully modeled using the dynamic mesh method. Additionally, a modified Wagner number (WaPR) is proposed to predict in advance the optimal conditions for achieving a more uniform deposit. This innovative approach contributes to the advancement of theoretical understanding and will improve practical applications in electrochemical deposition and dissolution processes.

Enhanced Medical Image Segmentation using Transfer Learning with Res101_UNet: Experimental Insights and Comparative Performance Analysis

Throughout the past few decades, artificial intelligence and machine learning have seen a lot of active research in areas such as computer vision, natural language processing, and speech processing. As a result, deep learning models became state-of-the-art for computer vision tasks such as object detection, classification, segmentation, and other allied tasks. Of course, the fruits of this research are extended to the design of robust and reliable digital health systems as well as other applications in the healthcare sector. Many clinical applications require the automatic segmentation of medical images. Recent deep learning-based approaches have demonstrated state-of-the-art performance in medical image segmentation tasks. In addition to their ability to automatically extract features and generalize over large amounts of data, transfer learning based deep learning models have proven to be handy for data scared areas like medical domains. In this research, we investigate and demonstrate the efficacy of a DCNN-based transfer learning model -Res101_Unet, which has been trained and/or fine-tuned to execute tumor tissue segmentation tasks in MRI, CT, PET, and X-RAY pictures of medical organ scans with little data. For our experimental study, we employed two image datasets: 'Liver Tumor' and 'Gland Colon Cancer', both obtained from the Kaggle portal. This experimental setup includes an Open-Source segmentation model API. Our findings indicate that domain similarity-based transfer learning can be used to data-scarce sectors. We achieved 98.47% accuracy and a IoU score of 0.9891 on Liver Tumor data and 0.6956 accuracy and a IoU score of 0.7043 on gland colon dataset.

Exploring EFL students’ prompt engineering in human–AI story writing: an activity theory perspective

ABSTRACT This study applies Activity Theory to investigate how English as a foreign language (EFL) students prompt generative artificial intelligence (AI) tools during short story writing. Sixty-seven Hong Kong secondary school students created their own generative-AI tools using open-source language models and wrote short stories with them. The study collected and analyzed the students’ generative-AI tools, short stories, and written reflections on their conditions or purposes for prompting. The research identified three main themes regarding the purposes for which students' prompt generative-AI tools during short story writing: a lack of awareness of purposes, overcoming writer's block, and developing, expanding, and improving the story. The study also identified common characteristics of students’ activity systems, including the sophistication of their generative-AI tools, the quality of their stories, and their school's overall academic achievement level, for their prompting of generative-AI tools for the three purposes during short story writing. The study's findings suggest that teachers should be aware of students’ purposes for prompting generative-AI tools to provide tailored instructions and scaffolded guidance. The findings may also help designers provide differentiated instructions for users at various levels of story development when using a generative-AI tool.

Coupling of photovoltaic thermal with hybrid forward osmosis-membrane distillation: Energy and water production dynamic analysis

This study proposes a novel solar photovoltaic thermal (PVT)-driven forward osmosis (FO) and membrane distillation (MD) system for desalinating brackish water, addressing water scarcity issues by combining the advantages of FO's low operating cost with MD's ability to treat hypersaline water. A physics-based dynamic model was developed for the system, and the model equations were discretized and solved using an open-source algebraic modeling language. Model validation involved verifying the consistency of outputs between successive unit operations and conducting sensitivity analysis by varying one operating condition at a time. The analysis considered varying weather conditions while maintaining constant operating conditions. Increasing the flow rate of brackish water (cooling fluid) from 19.79 to 59.38 kg/h improved thermal efficiency by 28.7 %; however, FO and MD water fluxes decreased by 1.12 LMH and 7.77 kg/m2/h, respectively. Peak thermal efficiency (81.2 %) was achieved at the highest flow rate. Incorporating heat transfer analysis into FO models goes beyond the common focus on mass transfer, allowing for a more comprehensive understanding of energy dynamics within the system. Both FO and MD water fluxes improved with irradiance and ambient temperature but decreased with cooling fluid flow rate and wind velocity. Increasing feed concentration to 10,000 mg/l reduced water flux by 0.66 LMH, and for each 0.5 M increase in draw solution concentration, FO water flux increased by >10 LMH. Compared to MD water flux, FO water flux is more sensitive to changes in concentration than to changes in weather conditions.

X-Phishing-Writer: A Framework for Cross-lingual Phishing E-mail Generation

Cybercrime is projected to cause annual business losses of $10.5 trillion by 2025, a significant concern given that a majority of security breaches are due to human errors, especially through phishing attacks. The rapid increase in daily identified phishing sites over the past decade underscores the pressing need to enhance defenses against such attacks. Social Engineering Drills (SEDs) are essential in raising awareness about phishing yet face challenges in creating effective and diverse phishing e-mail content. These challenges are exacerbated by the limited availability of public datasets and concerns over using external language models like ChatGPT for phishing e-mail generation. To address these issues, this article introduces X-Phishing-Writer, a novel cross-lingual Few-shot phishing e-mail generation framework. X-Phishing-Writer allows for the generation of e-mails based on minimal user input, leverages single-language datasets for multilingual e-mail generation, and is designed for internal deployment using a lightweight, open-source language model. Incorporating Adapters into an Encoder–Decoder architecture, X-Phishing-Writer marks a significant advancement in the field, demonstrating superior performance in generating phishing e-mails across 25 languages when compared to baseline models. Experimental results and real-world drills involving 1,682 users showcase a 17.67% e-mail open rate and a 13.33% hyperlink click-through rate, affirming the framework’s effectiveness and practicality in enhancing phishing awareness and defense.

Automatic quantification of scapular and glenoid morphology from CT scans using deep learning

ObjectivesTo develop and validate an open-source deep learning model for automatically quantifying scapular and glenoid morphology using CT images of normal subjects and patients with glenohumeral osteoarthritis. Materials and MethodsFirst, we used deep learning to segment the scapula from CT images and then to identify the location of 13 landmarks on the scapula, 9 of them to establish a coordinate system unaffected by osteoarthritis-related changes, and the remaining 4 landmarks on the glenoid cavity to determine the glenoid size and orientation in this scapular coordinate system. The glenoid version, glenoid inclination, critical shoulder angle, glenopolar angle, glenoid height, and glenoid width were subsequently measured in this coordinate system. A 5-fold cross-validation was performed to evaluate the performance of this approach on 60 normal/non-osteoarthritic and 56 pathological/osteoarthritic scapulae. ResultsThe Dice similarity coefficient between manual and automatic scapular segmentations exceeded 0.97 in both normal and pathological cases. The average error in automatic scapular and glenoid landmark positioning ranged between 1 and 2.5 mm and was comparable between the automatic method and human raters. The automatic method provided acceptable estimates of glenoid version (R2 = 0.95), glenoid inclination (R2 = 0.93), critical shoulder angle (R2 = 0.95), glenopolar angle (R2 = 0.90), glenoid height (R2 = 0.88) and width (R2 = 0.94). However, a significant difference was found for glenoid inclination between manual and automatic measurements (p < 0.001). ConclusionsThis open-source deep learning model enables the automatic quantification of scapular and glenoid morphology from CT scans of patients with glenohumeral osteoarthritis, with sufficient accuracy for clinical use.

GPT-4 performance on querying scientific publications: reproducibility, accuracy, and impact of an instruction sheet

BackgroundLarge language models (LLMs) that can efficiently screen and identify studies meeting specific criteria would streamline literature reviews. Additionally, those capable of extracting data from publications would enhance knowledge discovery by reducing the burden on human reviewers.MethodsWe created an automated pipeline utilizing OpenAI GPT-4 32 K API version “2023–05-15” to evaluate the accuracy of the LLM GPT-4 responses to queries about published papers on HIV drug resistance (HIVDR) with and without an instruction sheet. The instruction sheet contained specialized knowledge designed to assist a person trying to answer questions about an HIVDR paper. We designed 60 questions pertaining to HIVDR and created markdown versions of 60 published HIVDR papers in PubMed. We presented the 60 papers to GPT-4 in four configurations: (1) all 60 questions simultaneously; (2) all 60 questions simultaneously with the instruction sheet; (3) each of the 60 questions individually; and (4) each of the 60 questions individually with the instruction sheet.ResultsGPT-4 achieved a mean accuracy of 86.9% – 24.0% higher than when the answers to papers were permuted. The overall recall and precision were 72.5% and 87.4%, respectively. The standard deviation of three replicates for the 60 questions ranged from 0 to 5.3% with a median of 1.2%. The instruction sheet did not significantly increase GPT-4’s accuracy, recall, or precision. GPT-4 was more likely to provide false positive answers when the 60 questions were submitted individually compared to when they were submitted together.ConclusionsGPT-4 reproducibly answered 3600 questions about 60 papers on HIVDR with moderately high accuracy, recall, and precision. The instruction sheet's failure to improve these metrics suggests that more sophisticated approaches are necessary. Either enhanced prompt engineering or finetuning an open-source model could further improve an LLM's ability to answer questions about highly specialized HIVDR papers.

Evaluating Molecular Complexity with Open-Source Machine Learning Approaches to Predict Process Mass Intensity.

The application of green chemistry is critical for cultivating environmental responsibility and sustainable practices in pharmaceutical manufacturing. Process mass intensity (PMI) is a key metric that quantifies the resource efficiency of a manufacturing process, but determining what constitutes a successful PMI of a specific molecule is challenging. A recent approach correlated molecular features to a crowdsourced definition of molecular complexity to determine PMI targets. While recent machine learning tools show promise in predicting molecular complexity, a more extensive application could significantly optimize manufacturing processes. To this end, we refine and expand upon the SMART-PMI tool by Sheridan et al. to create an open-source model and application. Our solution emphasizes explainability and parsimony to facilitate a nuanced understanding of prediction and ensure informed decision-making. The resulting model uses four descriptors-the heteroatom count, stereocenter count, unique topological torsion, and connectivity index chi4n-to compute molecular complexity with a comparable 82.6% predictive accuracy and 0.349 RMSE. We develop a corresponding app that takes in structured data files (SDF) to rapidly quantify molecular complexity and provide a PMI target that can be used to drive process development activities. By integrating machine learning explainability and open-source accessibility, we provide flexible tools to advance the field of green chemistry and sustainable pharmaceutical manufacturing.

A Systematic Review and Comprehensive Analysis of Pioneering AI Chatbot Models from Education to Healthcare: ChatGPT, Bard, Llama, Ernie and Grok

AI chatbots have emerged as powerful tools for providing text-based solutions to a wide range of everyday challenges. Selecting the appropriate chatbot is crucial for optimising outcomes. This paper presents a comprehensive comparative analysis of five leading chatbots: ChatGPT, Bard, Llama, Ernie, and Grok. The analysis is based on a systematic review of 28 scholarly articles. The review indicates that ChatGPT, developed by OpenAI, excels in educational, medical, humanities, and writing applications but struggles with real-time data accuracy and lacks open-source flexibility. Bard, powered by Google, leverages real-time internet data for problem solving and shows potential in competitive quiz environments, albeit with performance variability and inconsistencies in responses. Llama, an open-source model from Meta, demonstrates significant promise in medical contexts, natural language processing, and personalised educational tools, yet it requires substantial computational resources. Ernie, developed by Baidu, specialises in Chinese language tasks, thus providing localised advantages that may not extend globally due to restrictive policies. Grok, developed by Xai and still in its early stages, shows promise in providing engaging, real-time interactions, humour, and mathematical reasoning capabilities, but its full potential remains to be evaluated through further development and empirical testing. The findings underscore the context-dependent utility of each model and the absence of a singularly superior chatbot. Future research should expand to include a wider range of fields, explore practical applications, and address concerns related to data privacy, ethics, security, and the responsible deployment of these technologies.

A comprehensive evaluation of large language models in mining gene relations and pathway knowledge.

Understanding complex biological pathways, including gene-gene interactions and gene regulatory networks, is critical for exploring disease mechanisms and drug development. Manual literature curation of biological pathways cannot keep up with the exponential growth of new discoveries in the literature. Large-scale language models (LLMs) trained on extensive text corpora contain rich biological information, and they can be mined as a biological knowledge graph. This study assesses 21 LLMs, including both application programming interface (API)-based models and open-source models in their capacities of retrieving biological knowledge. The evaluation focuses on predicting gene regulatory relations (activation, inhibition, and phosphorylation) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway components. Results indicated a significant disparity in model performance. API-based models GPT-4 and Claude-Pro showed superior performance, with an F1 score of 0.4448 and 0.4386 for the gene regulatory relation prediction, and a Jaccard similarity index of 0.2778 and 0.2657 for the KEGG pathway prediction, respectively. Open-source models lagged behind their API-based counterparts, whereas Falcon-180b and llama2-7b had the highest F1 scores of 0.2787 and 0.1923 in gene regulatory relations, respectively. The KEGG pathway recognition had a Jaccard similarity index of 0.2237 for Falcon-180b and 0.2207 for llama2-7b. Our study suggests that LLMs are informative in gene network analysis and pathway mapping, but their effectiveness varies, necessitating careful model selection. This work also provides a case study and insight into using LLMs das knowledge graphs. Our code is publicly available at the website of GitHub (Muh-aza).